Throughput Issues

Hey guys,

I’ve noticed the past few days i’m struggling to break ~30Mbps to my linux server, previously i was able to hit ~300Meg pretty easily.

Nothing changed my end, i was wondering if it was possible to check debug logs to see if anything stange is going on?

I’m using TS on my router (edgerouter 4) and this test was done from a windows 10 machine, both on the latest version.

Cheers!

[ 4] local 100.93.90.51 port 59878 connected to 100.124.64.14 port 5201
[ ID] Interval Transfer Bandwidth
[ 4] 0.00-1.01 sec 2.50 MBytes 20.7 Mbits/sec
[ 4] 1.01-2.01 sec 3.38 MBytes 28.4 Mbits/sec
[ 4] 2.01-3.01 sec 3.00 MBytes 25.1 Mbits/sec
[ 4] 3.01-4.01 sec 3.50 MBytes 29.4 Mbits/sec
[ 4] 4.01-5.01 sec 3.50 MBytes 29.4 Mbits/sec
[ 4] 5.01-6.00 sec 4.12 MBytes 34.8 Mbits/sec
[ 4] 6.00-7.00 sec 4.12 MBytes 34.6 Mbits/sec
[ 4] 7.00-8.00 sec 4.12 MBytes 34.6 Mbits/sec
[ 4] 8.00-9.01 sec 4.12 MBytes 34.3 Mbits/sec
[ 4] 9.01-10.00 sec 4.12 MBytes 34.9 Mbits/sec


[ ID] Interval Transfer Bandwidth
[ 4] 0.00-10.00 sec 36.5 MBytes 30.6 Mbits/sec sender
[ 4] 0.00-10.00 sec 36.5 MBytes 30.6 Mbits/sec receiver

iperf Done.

EDIT: both clients have a direct connection too, status from my external linux box

100.80.133.91 edgerouter-4 xxxxxxxxxxxxxxxx@ linux active; direct xxxxxxxxxx:1024, tx 1851972 rx 52234380

The main thing to check is whether it is deciding to go through a DERP relay. tailscale ping can show this:

$ tailscale ping server1
pong from server1 (100.99.98.97) via 10.1.10.13:40952 in 3ms

vs:

$ tailscale ping server2
pong from server2 (100.99.98.96) via DERP(fra) in 88ms

If it is using DERP, tailscale netcheck might provide clues as to why.

Thanks @DGentry, no DERP in use, however I believe i’ve tracked it down to something in my Edgerouter4, not entirely sure what yet as I have no QoS set up and CPU / RAM isnt maxing when running iperf so it’s a bit of a strange one!